Estimated probit models with financial crisis and the crisis: An evaluation on Turkey

There have been many financial crises since the beginning of the 1990s, and these crises have had negative consequences that have deeply affected the whole world. Especially the 1994 Mexican Crisis, the 1997-1998 Southeast Asian Crisis, the 1998 Russian Crisis, the 1998-1999 Brazilian Crisis and finally the 2008 Global Economic Crisis had devastating effects on economies both at the regional level and around the world. Turkey, on the other hand, experienced the 1994 crisis, November 2000 and February 2001 crises, and the country's economy was greatly affected by the 2008 Global Crisis. The increase in the number and severity of financial crises has led to the need to predict crises; however, it also increased the interest in the subject. For this reason, there are a number of theoretical and empirical studies in the literature aimed at determining the determinants of crises. Of course, predicting financial crises is a process that requires hard effort. However, past financial crises have shown two matters: first, to improve economic forecasting indicators, and second, to be able to develop an early warning system that works well for economic crises. In this context, economic forecast indicators are an extremely useful tool to prevent major catastrophes in the economy and to help policy makers make better decisions and prevent greater losses. The aim of the study created within the scope of this requirement is to present an empirical study for the prediction of financial crises with the probit model method for Turkey's 1992:M01-2019:M12 monthly and 1998:Q1-2019:Q4 quarter. In order to determine the financial crisis periods, the financial pressure index was created and used as a dependent variable. The independent variables included in the analysis were carefully selected as a result of the literature review, and it was tested whether the variables could be a leading indicator for the financial crises in Turkey. From the monthly data included in the model; there is a statistically significant relationship between current account balance, overnight borrowing interest rate, Central Bank reserves, composite leading indicators, foreign trade balance, stock price index, inflation rate variables and dependent variable. From the quarterly data; it was concluded that there is a statistically significant relationship between inflation rate, export coverage ratio, M2 money supply to GDP ratio, Central Bank reserves, short-term external debt ratio, foreign direct investments to GDP, total credit volume variables and the dependent variable.